© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112048
    Prev
    Next
    INDEX
    Explanations

    customized, multifaceted, or specific descriptions

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron responds to subword tokens that end with an underscore (e.g. “friendly_”, “default_”, “invalid_”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    皷
    -2.36
     fünf
    -2.19
    xvii
    -2.11
     terse
    -1.98
     praktische
    -1.97
     Atención
    -1.96
    provides
    -1.96
     Però
    -1.95
     Quando
    -1.93
     ¨
    -1.93
    POSITIVE LOGITS
    且
    2.28
     enhances
    2.17
    !!!!!
    2.14
    just
    2.14
    かつ
    2.02
     customized
    2.00
     multifaceted
    1.99
     undeniably
    1.95
    {~
    1.91
    !!!!
    1.91
    Activations Density 0.068%

    No Known Activations